There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Wireless indoor positioning systems have become very popular in recent years. These systems have been successfully used in many applications such as asset tracking and inventory management. This paper provides an overview of the existing wireless indoor positioning solutions and attempts to classify different techniques and systems. Three typical location estimation schemes of triangulation, scene analysis, and proximity are analyzed. We also discuss location fingerprinting in detail since it is used in most current system or solutions. We then examine a set of properties by which location systems are evaluated, and apply this evaluation method to survey a number of existing systems. Comprehensive performance comparisons including accuracy, precision, complexity, scalability, robustness, and cost are presented.

Survey of Wireless Indoor Positioning Techniques and Systems

The PASCAL Visual Object Classes Challenge

An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

For calculating SAD values for two regular groups of multi-bit values, such as macroblocks of different video frames, there are calculated (602) differences of multi-bit values taken in pairs from corresponding locations of said two regular groups. Thus we produce uncorrected absolute differences and correction bits, which are compressed (611) to a SAD. As a part of said compressing (611), correction bits are compressed (613) initially separately from the compression (612) of uncorrected absolute differences. The compressed correction bits are fed into a number of locations in the main process of compressing (612) absolute differences.

Circuit and method for computing a sum of absolute differences fast and effectively

Mapping of Kanerva’s Sparse Distributed Memory (SDM) is presented for a partial tree shape neurocomputer. We discuss the original model and give a detailed parallel mapping for improved SDM models, using Karlsson’s activation mechanism and efficient reading methods. Both row-wise and column-wise mappings can be effectively used, but column-wise mapping is analysed in more details because it minimizes communication in the improved read operation. Instead, rowwise mapping parallelized the activation mechanism so that it is better in slower activation methods or when needed a very fast writing procedure. Performance analysis deals with SDM matrix sizes, memory capacity and load balance.

Sparse Distributed Memory Mapping on Partial Tree Shape Neurocomputer

Modern ASIC and FPGA based embedded products use model based design, in which both hardware and software are developed in parallel. Previously HW was completed first and the information handed over to SW team, typically in the form of register tables. The information was even manually copied to SW code, making any changes error-prone and laborious. IP-XACT is the most feasible standard to model HW also for the SW needs. The HW design connectivity and overall memory layout may change due to component instantiations, configurations and conditional operation states, which makes it difficult to create register tables even for documentation. Current register design tools fall short in serving the needs of both the HW and SW designer. All of them rely on low-level IP-XACT XML description that leaves much of the interpretation responsibility to the user, and most do not have any user-friendly visualization of the whole. Our approach is to perform a design connectivity analysis through hierarchies and configurations, analyze the memory structure based on it, and finally provide a novel, interactive visualization of system-wide memory maps. Our new Memory Designer for Kactus2 serves the hardware design by accessing and adjusting the memory layout from a single view disclosing end-to-end paths between masters and slaves. For SW development the Memory Designer shows the "programmer's view" to the system by hiding intermediate address translations and other connectivity related issues. As far as we know, this is the first visual, user-friendly memory tool for model based embedded system design.

Analysis and Visualization of Product Memory Layout in IP-XACT

A Statistical Model of Spine Shape and Material for Population-Oriented Biomechanical Simulation

In this work, the problem of efficient configuration and management of adaptive routers is considered. Adaptive routers use intellectual models to change dynamically the parameters of the queuing policies based on various criteria. By now, the policy-based information model can be used to provide configuration data for these routers. However, the existent scheme fails to provide a set of classes for adaptive queuing models. Thus, a creation of an extension to the policy-based information model is considered and a set of possible classes is highlighted. Furthermore, an example is given that considers how parameters of an adaptive queuing model can be represented as a set of policy classes and used within the DiffServ framework.

Timo Hämäläinen

Papers

Circuit and method for computing a sum of absolute differences fast and effectively

Sparse Distributed Memory Mapping on Partial Tree Shape Neurocomputer

Analysis and Visualization of Product Memory Layout in IP-XACT

A Statistical Model of Spine Shape and Material for Population-Oriented Biomechanical Simulation

Extension of the QoS policy information model to support adaptive queuing techniques